dxflow Python SDK

dxflow Logo

dxflow Python SDK#

dxflow is a powerful, domain-agnostic platform for executing, managing, and productionizing scientific models, from bioinformatics and cheminformatics to computational fluid dynamics and digital physics.

Built by the DiPhyX team, dxflow is engineered to simplify the entire lifecycle of scientific workflows, making compute-intensive research reproducible, scalable, and accessible, whether on local machines or in the cloud.

Key Features#

  • Job Execution & Management (Cloud + Hybrid): Run scientific models on DiPhyX-managed cloud infrastructure or connect your own compute resources in a hybrid environment. dxflow supports dynamic integration of HPC clusters, on-prem systems, or third-party cloud environments to provide seamless execution and job management.

  • Experiment Tracking: Group jobs into experiments to compare configurations, costs, and results.

  • Reproducibility First: Every job run with dxflow is versioned, documented, and fully reproducible.

  • Production-Ready Pipelines: Move scientific models from prototypes to scalable, production-grade executions with ease.

  • Integrated Post-Processing: Visualize results or run analyses through integrated Jupyter notebooks.

  • Robust API Access: Use the REST API and Python SDK to automate, extend, or integrate with external systems.

  • Command-Line Interface (CLI): Run and manage jobs via the CLI for a developer-friendly workflow.

  • Version Control and Documentation: dxflow automatically tracks code, configurations, inputs, and results for every job.

  • Cost and Resource Optimization: Run cost-effective simulations with built-in resource tracking and environment selection.

Designed For#

  • Scientific Researchers

  • Engineers and Simulation Experts

  • Bioinformaticians & Cheminformaticians

  • AI/ML Researchers applying models to physical or biological systems

Supported Domains#

dxflow supports compute-intensive scientific workflows across:

  • Bioinformatics/Cheminformatics: e.g., GROMACS, AMBER

  • CFD (Computational Fluid Dynamics): e.g., OpenFOAM, ANSYS Fluent, SU2

  • Physics & Engineering Simulations

  • Custom ML/AI Pipelines with physical models

How It Works#

  1. Define your model and environment

  2. Use dxflow CLI or Python SDK to launch jobs

  3. Monitor, adjust, and post-process in real time

  4. Archive and share reproducible results

Installation#

To install dxflow, you can use pip:

pip install dxflow

License#

Proprietary – © DiPhyx Inc.

More Info#

Contents